% Seasonal Respome of Fiona; Me yooii? ‘to Iierbicids cooperation wit CONTENTS SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . ............ .. MATERIALS AND METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . l Experimental Site and Layout Plot . . . . . . . . . . . . . . . . .. Chemical Applications and Control Ratings . . . . . . . . Plant Characteristics . . . . . . . . . . . . . . . . . , . . . . . . . . . . . Environmental Variables . . . . . . . . . . . . . . . . . . . . . . . . . . Statistical Analyses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . .. Plant Response to Herbicides . . . . . . . . . . . . . . . . . . . . . Plant Characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Environmental Variables . . . . . . . . . . . . . . . . . . . . . . . . . . . Correlation and Regression Analyses . . . . . . . . . . . . . . . LITERATURE CITED . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . APPENDIX . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . SUMMARY Honey mesquite (Prosopis juliflora (Swartz) DC. var. glai (Torr.) Cockerell) was sprayed with 0.56 + 0.56 kilogram per (kg/ha) of picloram (4-amino-3,5,6-trichloropicolinic acid) j (2,4,5-trichlorophenoxy) acetic acid) or 1.12 kg/ha of 2,4, dates during 1972 and 1973. Honey mesquite was most - g controlled during late April, May, and June. With this and 2-year study, at the 17 best dates, picloram + 2,4,5-T alonej the canopy 92 percent and killed 59 percent of the plan’ 2,4,5-T was less effective, reducing the canopy 66 percent . _ 5 percent of the plants. Over 36 dates from March 24 to O during a 4-year period, percent honey mesquite canopy was directly correlated with total phloem thickness, rat? xylem ring radial growth, and rate of upward methylene o if ment in the xylem, and was inversely correlated with mini moisture stress. In simple regression equations, for all 36d March 24 to October 9, rate of new xylem radial growth‘ best predictive equation. For 33 dates between April 29 an; 31, the best equation contained soil moisture level at a df centimeters (cm), minimum leaf moisture stress, or rate o_ methylene blue dye movement. At 13 increasingly effecrf season dates, control was best predicted by soil temper depth of 91 cm. At 22 decreasingly effective summer dates} was best predicted by minimum leaf moisture stress for canopy reduction by both herbicides and by rate of new i, radial growth for percent plants killed by picloram + 2,4,5- ‘ new xylem ring radial growth and thickness of translocatin were the factors appearing most often in the equations. ‘ ‘See Appendix for equivalent English units. Honey mesquite varies widely in its response to icides. In some cases almost all plants are killed ' given treatment; at other times very few plants 5 illed. In west Texas, when growing conditions avorable, 0.56 kilogram per hectare (kg/ha) of -T generally destroys most top growth and kills Vt25 percent of the plants (5). The most effective _, ents have occurred 50 to 80 days after the first is appeared in the spring when the leaves were formed and dark green. Treatments have been ective when applied before this time or during ‘ er and fall when the plant was not actively 1 mg. IDahl et al. (2) found that soil temperature of 27° d above at the 46-centimeter (cm) depth was the t important factor affecting the response of mesquite to 2,4,5-T. No plants were killed A soil temperature was in the low 20's ‘C or be- Plants most easily killed were those having ma- f, dark-green foliage and mature legumes. Trees pland and sandy soils were apparently more ptiblelto 2,4,5-T than were those on bottom- ‘and clay sites because of the difference in soil rature. _obison, Fisher, and Cross’ and Fisher et al. (6) j shown that honey mesquite is more suscepti- o mixtures of picloram + 2,4,5-T than 2,4,5-T In six ranch tests, picloram + 2,4,5-T at 0.28 + rkg/ha killed an average of 52 percent of the s compared to 21 percent for 0.56 kg/ha of -T alone. Thus, factors were present that pre- d the death of all plants. Meyer et al. (10) found to E. D., C. E. Fisher, and B. T. Cross. 1970. Control of uite and associated west Texas brush with 2,4,5-T/picloram inations. Proc. South. Weed Sci. Soc. 232219 (Abstr.). k physiologist, Agricultural Research Service, U. S. Depart- y of Agriculture, The'\Texas Agricultural Experiment Station artment of Range Science). _'0n of a trademark or a proprietary product does not consti- v guarantee or a warranty of the product by The Texas Ag- ral Experiment Station or the U. S. Department of Agricul- nd does not-imply its approval to the exclusion of other cts that also may be suitable. amt lwvnx 6f cg Mesquite to Iicrbicidcs R. E. Meyer* that picloram alone and picloram + 2,4,5-T (1:1) at 0.56- and 1.12-kg/ha rates were equally effective and more effective than 2,4,5-T at the same rates. Meyer et al. (11) showed that the toxic agents from 2,4,5-T, picloram, and picloram + 2,4,5-T sprays were translocated from the leaves to the stem of honey mesquite within 4 days after application. Brady (1) found herbicide applications in May to be more effective than those made later in the growing season on sweetgum (Liquidambar styracif/ua L.), green ash (Fraxinus pennsy/vanica Marsh.), and water oak (Quercus nigra L.). Effective killing of plant tops was attributed to the high rates of absorp- tion and translocation of the herbicide. A 4-day ab- sorption period was closely correlated with the con- trol of top growth 1 year later. Davis et al. (3) sprayed honey mesquite with 0.56- and 1.12-kg/ha per acre rates of 2,4,5-T, picloram, and a mixture of picloram + 2,4,5-T. After 48 hours, highest concentrations of herbicides occurred in the phloem of those plants sprayed in June and lowest in those sprayed in Au- gust. Similar levels of 2,4,5-T occurred in the phloem from application of either 0.56 or 1.12 kg/ha, but more than three times as much picloram occurred in plants sprayed with 1.12 kg/ha as in those sprayed with 0.56 kg/ha. At College Station, Texas, honey mesquite leaves begin emerging about the end of March (13). Emergence is probably controlled by environmental conditions of high temperatures (9). The new stems elongate for approximately 1 month beginning about the first of April and continuing until tip ab- oration occurs (13). Then the plant enlarges radially in May and June, producing new translocating phloem and a new xylem growth ring. Fisher, Fults, and Hopp (4) and Meyer, Haas, and Morton’ showed that upward translocation of dye occurred in the new, outermost xylem ring of honey mesquite; the dye streak widened from about 0.4 cm at the point of 3Meyer, R. E., R. H. Haas, and H. L. Morton. 1965. Mesquite stem, its structure, seasonal growth characteristics, and area of active xylem dye movement. Proc. South. Weed Sci. Soc. Conf. 18:632 (Abstr.). injection t0 about a 5-cm width at 1.2 meter (m) and moved about 52 cm per hour. Meyer, Haas, and Wendt (12), using greenhouse plants under controlled conditions, showed that either low soil temperature (13° C) or cool aerial environment (11° to 25° C) retarded shoot growth; however, maximum shoot growth occurred as a re- l sult of an interaction of optimum soil temperature (29° C) and aerial environment of 20.9 millimeters (mm) vapor pressure deficit (23° to 40° C). Several workers studied the total available car- bohydrate level in honey mesquite stems and roots (5, 14, 15). Total available carbohydrate level was lowest in May. At this time the partial drain of food reserves was used for leaf and floral production and radial enlargement of stems and roots. Haas and Dodd (8) studied water stress of the honey mesquite leaf petiole. They showed a diurnal pattern of low stress at pre-dawn and post-sunset periods and high stress during the day. A gradual increase in stress occurred at all three periods throughout the season when leaves were present. Meyer et al. (10) found similar results in untreated plants; herbicides slightly reduced stress levels dur- ing the maximum stress periods for 2 or 3 days be- fore the leaf was killed. Meyer et al. (10) found that most effective con- trol of honey mesquite occurred from treatments with picloram, picloram + 2,4,5-T, and 2,4,5-T applied between April 30 and July 6. Generally, thickness of translocating phloem, rate of upward dye movement in the xylem, lower minimum rela- tive humidity, high soil moisture, and higher rainfall before spraying were directly correlated with higher plant control, while measurements of new xylem thickness, air temperature, maximum relative humidity before spraying, maximum soil tempera- ture, rainfall after spraying, and leaf moisture stress were inversely correlated with high percentage con- trol by herbicide treatments. The objectives of this study were to develop a reliable means of estimating the ultimate response of honey mesquite to herbicides and to determine the interrelationships of various plant and environ- mental variables. This study is similar to that of Meyer et al. (10); major differences are that this was conducted on a different site, the number of chemi- cal treatments was reduced from six to two, the number of dates sprayed was increased from 14 to 22, and the percentage of spraying dates was in- creased during the May, June, and July period. Ap- propriate data from both experiments were com- bined. MATERIALS and METHODS Experimental Site and Plot Layout A 13-ha site near Millican, Texas, with a dense stand of honey mesquite plants 1.2 to 2 m tall was 4 selected. Most honey mesquite plants had thl five stems that had emerged near the base i plant. The area was an upland site with a" 3-percent slope. The soil was a Wilson clay j About 1,200 plants were tagged in groups o, with at least a 1-m space between plants from a ing plots. Five replications of plants were us each of two treatments at each of 22 dates li _ Table 1 during the 1972 and 1973 period. A f design was used. ChemicalApplications and Control Ratings The herbicide treatments included a l, 0.56-kg/ha rate of picloram + 2,4,5-T and kg/ha rate of 2,4,5-T alone. The potassium j picloram and the propylene glycol butyl ethe of 2,4,5-T were applied in water containing ll cent volume per volume (v/v) of the surfacta tifilm X-77, containing alkylarylpolyoxyet glycols, free fatty acids, and isopropanol. a The herbicides were applied either in t , ning or early morning with a hand-carried c0 sed air, three-nozzle boom sprayer. The her were applied at a spray volume of 187 liters tare (L/ha). g Visual ratings of percent canopy red " j which measure the amount of stem tissue l; and percent dead plants were made July 31f and May 28, 1974, the season following spra E Plant Characteristics New stem length was measured at weekl vals after the onset of elongation growth in; Plant characteristics measured at each sprayiij included transectional dimensions of stem f the day of spraying, rate of upward move l. dye in the stem xylem within a day of sprayi cent total available carbohydrates in the st day of spraying, and minimum and maxim, moisture stress within a day of spraying. ' Unsprayed plants in the experimental f‘ used for all measurements. For the most vf same methods were used in this study as Meyer et al. (10). Five new stems were ;{ each of five trees and measured at weekly l until May, when elongation growth had-ts Stem tissue transectional dimensions we su red from pieces of stem cut 15 to 30 cm L. ground from 10 trees. Thicknesses of the p total and translocating phloem, and xylem< rings were measured. Rate of new xylem ri growth during the 2-week period before f was calculated from the total new ring t measurements taken during the growing se Rate of upward methylene blue dye cf was determined in 20 trees at each date. A f, cent aqueous methylene blue dye solution l fused into the stem xylem from a transfusiof ipped with a N0. 16 hypodermic needle. The _les were inserted under the bark between 9:00 .- 9230 a.m. CST and removed 30 minutes later. * rate is presented as cm/hr. ‘Percent total available carbohydrates was de- . ined from stem samples cut 15 to 3O cm above ground from 5 to 10 trees. The bark was peeled » xylem segments from the outermost 1 to 2 mm T. lem were ground to pass a 23.6-mesh/cm sieve. pectrophotometric method was used (15). The t tissue was digested in 0.2 N HC1 for 2 hours, mixture was filtered through activated charcoal, ‘een color was developed by adding an anthrone ent, the solution was heated to 97° C for 10 min- l, and the light absorption of the cooled solution read at 612 millimicrons (m,u.). Samples collected 969 and 1970 by Meyer et al. (10) were also lyzed by this method, and the results were used statistical computations. "Moisture stress in the leaves was determined a Scholander pressure apparatus (8). Moisture ss was determined in 20 mature (when available) es collected from four or five plants. The imum readings were made before dawn at 4 to 5 . CST, and the maximum readings were taken ‘i to 11:30 a.m. ironmental Variables Environmental variables measured included n air temperature the day of spraying and the ek period before spraying; mean soil tempera- " at depths of 30 and 91 cm the week of spraying} v ent soil moisture at depths of 0 t0 30 (hereafter 731 to 61 (hereafter 61), and 62 to 91 (hereafter 91) _' he week of spraying; and rainfall the 1-week and [nth periods before spraying. . Mean air temperatures were recorded at the site - a hygrothermograph. Soil temperatures were Ln with a recording soil thermograph. Soil ture was determined weekly using gravimetric ysis; five cores were dug with a screw-type ‘r (7). Rainfall data collected at College Station, j_ t 26 kilometers (km) away, were used for the rmination of rainfall. stical Analyses An analysis of variance and Duncan's multiple '_ test were calculated to determine significance ercent canopy reduction and percent dead s using a factorial design. A complete set of simple correlations was calcu- l for all combinations of herbicide treatments + 0.56 kg/ha of picloram + 2,4,5-T and 1.12 a of 2,4,5-T), = plant characteristics, and onmental variables in this and in the 1969 and §study by Meyer et al. (10). egressions were calculated to develop an indi- for predicting plant response to herbicides at f given date with plant characters and environmental factors. Also, the number of days from January 1 to the spraying date (T) and the time period (T) squared (T2) factors were added to ac- count for the curvilinear response of honey mes- quite to herbicides with time. A maximum R’ im- provement regression analysis was used. Equations were calculated with the best one- and three-plant and environmental variables. Means for all treat- ment replicates were used for herbicide responses in all correlation and regression analyses. RESULTS and DlSCUSSlON Plant Response to Herbicides Within 4 days after spraying, most leaves of plants sprayed with picloram + 2,4,5-T had remained intact, turned brown, and died. By 7 days, all leaves were dead. Leaves of plants sprayed with 2,4,5-T turned yellowish-green by the end of the third day. Some defoliation occurred, with the leaflets first de- taching from the rachises. By the end of 7 days, the remaining portions of the leaves were dead. Death .1 of the foliage generally occurred a day or two sooner when the plants were sprayed in late summer than in April or May. ln this study, picloram + 2,4,5-T at 0.56 + 0.56 kg/ha caused more canopy reduction and killed more plants than 2,4,5-T at 1.12 kg/ha (Table 1). In 1972, picloram + 2,4,5-T reduced the plant canopy most from April 29 to June 5, while in 1973 the mix- ture reduced plant canopy most from May 8 through June 22. The period for largest percentage of plants killed occurred during a slightly shorter interval than for most effective canopy reduction. The 2,4,5-T treatment caused the greatest amount of canopy re- duction on May 21 in 1972 and during the period of May 16 through lune 22 in 1973. Thus, both treat- ments were most effective in late April, May, and June. This period corresponds to the recommended 50 to 80 days after bud break for spraying honey mesquite. The 2,4,5-T treatment killed no significant numbers of honey mesquite plants at any date in either year. These results are similar to those in a 1969 and 1970 study by Meyer et al. (10). Plant con- trol was slightly greater in this study because a greater proportion of the applications was made during the susceptible treatment period. Figure 1 shows the seasonal response of honey mesquite during the 4-year period (1969, 1970,1972, and 1973). The data are plotted as the means of all treatments applied during the first and last half of each month. Plant control attains maximum effec- tiveness in late April, remains more or less constant during May and June, and then decreases rather rapidly later in the season. This general seasonal re- sponse curve is well known; however, on lighter soils, the 2,4,5-T treatment usually kills a significant number of plants. Table 1. Percent canopy reduction and dead honey mesquite sprayed with two herbicides at 22 dates during 1972 and 1973 at Millican, Texas1 Picloram + 2,4,5-T 2,4,5-T Date Canopy Dead Canopy Dead of reduction plants reduction plants spraying (%l l%l l%l l%l 1972 April 29 92 abcd 70 abc 55 klm 10 hi May 8 96 ab 80 ab 61 jk 8 hi May 14 95 ab 72 abc 62 iJk 4 hi May 21 97 a 72 abc 76 fgh 4 hi June 5 93 abcd 56 bcd 55 klm 0 i June 12 81 def 36 defg 56 klm Oi June 19 81 def 24 efghi 53 klm Oi July 6 79 efg 36 defg 55 klm O i . July 19 72 fghi] 16 fghi 41 n Oi July 31 44 lmn 8 hi 37 n Oi 1973 May 8 94 abc 56 bcd 62 ijk Oi May 16 97 a 84 a 75 fghi 8 hi May 29 94 abc 72 abc 78 efgh 12 ghi June 7 84 bcdef 48 cde 67 ghijk 4 hi June 2O 97 a 84 a 66 ghijk 4 hi June 22 89 abcde 40 def 66 ghijk O i June 26 81 def 2O fghi 62 jk Oi July 4 78 efgh 16 fghi 57 kl Oi July 12 82 cdef 28 efgh 44 lmn 0 i July 23 56 klm Oi 33 np Oi August 13 21 q 0i 23 pq 0i August31 18G 0i 17G 0i Mean 78 x 42 s 55 y 2 t 1Values in columns for canopy reduction of both herbicides, dead plants for both herbicides, and in the horizontal line for mean of canopy reduction for both herbicides or dead plants for both herb- icides not followed by the same letter are significantly different at the 5% level using the Duncan's multiple range test. Ratings were made July 31, 1973, and May 28, 1974, for the 1972 and 1973 treatments, respectively. Plant Characteristics New stems were initiated between March 15 and April 1, and they elongated rapidly until about April 30. In 1972, the 25 new stems averaged 30 cm in length by April 29 and showed no further growth. The 25 stems tagged in 1973 averaged 18 cm by May 3 and grew no further. The stem tips died and aborted after elongation ceased. No new stems were pro- duced during either year. Thus, stem elongation al- most ceased before maximum plant control was ob- tained. Stem transections measured at each spraying date were 15 to 21 mm in diameter. Annual radial stem growth occurred largely in the xylem. The new xylem ring began developing about the middle of April and attained a maximum thickness of 1.64 mm on July 31,1972, and 1.58 mm on July 12, 1973 (Table 2). The xylem ring essentially ceased enlarging ra- dially in the middle of July. 6 The other stern tissue thicknesses varie what; however, they fell in the following i‘ periderm — 0.16 to 0.29 mm; total phloem —l 0.87 mm; translocating phloem — 0.12 to 0. ‘ and mean for xylem rings other than that pr the current year — 1.12 cm in 1972 and 1.71] 1973. 4 Upward movement of methylene blue =3 curred almost entirely in the outermost ring.The most rapid dye movement occurred. May when the leaves had just fully mature 2). The dye movement during this period w ally greater in 1973 than either in 1972 or in 1970 (10). Warm mornings with full sunli adequate soil moisture seemed to promotet rapid dye translocation. Total available carbohydrates varied fro j 25.8 percent (Table 2). In 1972, the total carbohydrate level was low only on April 29; the carbohydrate level increased progressiv May 8 to a maximum on June 20 and 22. The 7_ percent total available carbohydrate level stem samples collected in 1969 and 1970 (y; assayed using the spectrophotometric met l the following: March, 27.7 percent; April, . cent; May, 18.4 percent; June, 28.4 perch 22.3 percent; August, 21.9 percent; and 21.6 percent. Normally the carbohydrate le y, only during the period of abundant leaf, fl0 i fruit production and elongation and radial! ment of the stem. In this study the low l; curred either before or during the early p best spraying period. i Moisture stress levels in leaves have. cycle (8, 10), being at a minimum at night a ing a maximum during late morning throu; ternoon. Minimum (pre-dawn) moisture st’ lowest at the earliest spraying dates and if gradually during late April, May and June, T‘ accelerated in July and August (Table 2). ‘l stress varied from 6.9 to 13.8 bars. Maxi ' moisture stress was normally two to thr’ higher than minimum stress. Maximum str increased rapidly from late April to the i May and then increased only slightly there; levels varied from 15.8 to 32.4 bars. . Environmental Variables Over both years, the range of mean air; ture the 1-week period before spraying du May, and June varied from 19° to 28° C i Temperature increased progressively at 4; May and June. Soil temperature at 91 cm increased i; 20° to 23° C in early May to 26° C by May;- High temperatures of 28° or 29° C were first‘ June 5 to July 12. No soil temperature a if occurred through August 31. In this study," perature at the 30 cm depth (data not s i 2. Plant characteristics at 22 dates of spraying honey mesquite at Millican, Texas Phloem _ I Rate of Leaf moisture thickness Xylem ring thickness upward Total availabm stress Trans- Total for Rate of dye stern Midday Total locating year growth movement carbohydrates Predawn maximum sprayed (mm) (mm) (mm) (mm/2 wk) (cm/hr) (%) (bars) (bars) 1972 0.53 0.12 0.19 0.19 143 18.2 8.4 15.8 .65 .14 .75 .21 192 23.5 9.0 24.7 .67 .13 .71 .25 180 23.1 9.1 24.0 .87 .25 .63 .25 169 25.1 8.5 25.7 .63 .18 .75 .24 92 24.2 9.7 28.9 .74 .22 1.18 .25 122 23.6 8.7 23.5 .65 .16 1.29 .25 148 23.2 9.4 25.9 .67 .29 1.32 .25 43 24.5 12.8 30.3 .71 .34 1.59 .25 105 25.6 11-1 28.1 .57 .29 154 .05 67 25.6 11.8 32.4 1973 .61 .37 .16 .16 328 17.0 7.7 19.2 .65 .25 .34 .34 332 17.2 7.2 17.7 .70 .21 .49 .29 307 18.4 10.4 24.2 .54 .19 .89 .30 157 21.0 8.5 24.2 .61 .24 1.06 .30 113 25.8 6.9 22.2 .61 .24 1.06 .30 113 25.8 6.9 22.2 .61 .23 1.20 .30 79 25.5 9.3 22.4 .79 .26 .97 .23 113 25.3 6.9 25.9 .55 .24 1.58 .08 88 23.0 9.3 22.3 .66 .26 1.22 .07 68 24.4 10.8 26.9 .69 .29 1.55 .07 55 24.4 12.6 26.8 .58 .24 1.21 .07 31 25.6 13.8 26.1 °__.°_,¢O-__°___°___° f ‘~, P |c LO RAM M \ \ \ \ -- m AR APR TMAY ‘ JU ‘ JUL 113m“ + 2,4,s-r % C.R. o---o % 0.2 ~---- O 2,4,s-r % c.|z. x-X o/o D.P. 4""4 Figure 1. Influence of 0.56 + 0.56 kg/ha of picloram + 2,4,5-T and 1.12 kg/ha of 2,4,5-T on percent canopy reduction (% C.R.) and percent dead honey mes- quite plants (% D.P.) treated on 36 dates during 1969, 1970, 1972, and 1973 at Bryan and Millican, Texas. 7 Table 3. Environmental ‘conditions at 22 dates of spraying honey mesquite at Millican, Texas1 . . Soil moisture 1 wk Mean air temperature Soil temperature before 1 wk before spraying at 91 cm3 30 cm 91 cm spraying Date sprayed (QC) (QC) (%) l%) (cm) 1972 . April 29 21 22 15 18 2.2 May 8 22 23 20 21 2.6 May 14 21 24 22 22 11.6 May 21 23 26 18 21 2.5 June 5 24 28 1 1 17 .0 June 12 25 28 23 24 .7 June 19 27 28 19 19 8.5 July 6 27 29 17 18 .2 July 19 28 28 15 16 .5 July 31 29 28 14 14 1 1973 May 8 19 20 21 23 3.6 May 16 21 23 21 23 1.3 May 29 25 26 13 22 .2 June 7 26 27 19 22 6.9 June 20 28 26 22 22 .3 June 22 27 26 22 22 i .6 June 26 25 26 18 22 6 July 4 28 27 14 21 .0 July 12 28 28 16 20 1.5 July 23 29 28 12 18 0 August 13 28 29 12 16 5 August 31 28 28 12 14 1 A Rain 1Soil temperature and moisture were measured the week of spraying. Air temperature was measured the day and 1-week period? ing. Rainfall was recorded the 1-week and 1-month period before spraying. 2Mean air temperature the day of spraying was no more than 3 degrees different from the mean of the 1-week period before spr I) 3Soil temperature at 30 cm did not vary from soil temperature at 91 cm by more than 2°C at any date. A mained within 2° C of that at 91 cm, whereas in the 1969 and 1970 study (10) the temperature at a depth of 91 cm was 3° to 6° C cooler than at 30 cm. Mean soil moisture levels over the 2-year period were 17, 19, and 20 percent at depths of 30, 61, and 91 cm, respectively (Table 3). Soil moisture at the 61-cm depth (data not shown) was within 3 percent of that at 91 cm. The period of highest moisture con- tent occurred generally in May and June. In 1972, however, low soil moisture occurred on April 29 and June 5, as well as in July. The lowest soil moisture occurred during late July and in August in 1973. The highest and lowest soil moisture levels were 25 per- cent at 61 cm on June 12 and 11 percent at 30 cm on June 5 in 1972. Rainfall was erratic, but most fell in the periods prior to the May and June spraying (Ta- ble 3). Table 4 shows the mean and range of all var- iables during the best 17 spraying dates from April 29 to July 6 over the 4-year period. All factors varied widely, and only soil temperature at the depth of 91 8 a cm and soil moisture at depths of 61 and significantly correlated with canopy re, plants treated with picloram + 2,4,5-T. ; were significantly correlated with 2,4,5-T dead plants with either treatment. a Correlation and Regression Analyses Correlations and regressions were for four seasonal intervals using data coll i; 4 years. First, all 36 dates covered the p' the onset of growth in March to Octobe‘ natural defoliation had become prominen, the 33 dates between April 29 and Augustl the period of almost full foliage. Third, 131 tween March 24 and May 29 represent the? increasing and early maximum effective s: Fourth, 22 dates from May 21 through ‘i. represent the period of decreasing herbip tiveness. Because insignificant numbers g A Means and range of honey mesquite control, plant charac- 3 and environmental variables at the 17 best spraying dates .’= from April 29 to July 6 during 1969, 1970, 1972, and _ Mean Minimum Maximum " canopy reduction 66 50 78 f dead plants 5 0 15 + 2,4,5-T, % canopy ion 92 77 97 } + 2,4,5-T, % dead V, . 59 15 84 ‘ Methylene dye _ ent, cm/hr 155 28 332 dsture stress (bars) 8.5 6.5 11.1 ' um 24.1 13.6 30.2 A thickness, mm 1 .59 .37 .87 'ocating .18 .08 .37 l’ ring thickness, mm ‘ .74 .16 1 .59 during 2-wk period 1 ore spraying .25 .10 .36 y ilable carbohydrates, % 21.8 12.0 30.4 ‘r temperature, °C of spraying 23.9 19.0 32.0 -period before spraying 23.7 19.0 30.0 . perature, 0C A deep 25.7 21.0 32.0 -- deep 23.2 17.0 28.0 l, cm _ period before spraying 2.81 .00 11.60 _- nth period before raying 111.20 .60 22.30 isture, % i 3V0 cm deep 18.2 8.0 30.01 61 cm deep 21 1 16.0 27.0 w 91 cm deep 21 :5 17.0 25.0 ; killed, no correlations or regressions are pre- w for percent plants killed by 2,4,5-T. _ oney mesquite control with herbicides was lated with plant and environmental variables. l- 36 dates (data not shown), R’ significance at % level = :0.32/:0.42, only total phloem ness (0.34 to 0.43), rate of new xylem ring h (0.62 to 0.72), and upward methylene blue movement (0.50 to 0.60) were directly corre- , and minimum leaf moisture stress (—0.35 to 1) was inversely correlated with percent canopy ction. The two best correlated factors, rate of xylem ring growth and upward movement of ylene blue dye, both formed parabolic curves ar to plant control (Figure 1) with time. or the 33 dates from April 29 through August tvering the period of almost full foliage, 13 var- were significantly correlated with plant con- able 5). Rates of new xylem ring radial growth ' pward movement of methylene blue dye again , positively correlated with control. The other rs more or less reflected the influence of a large number of treatments at the best to decreasingly effective dates. Rainfall and soil moisture levels at all three depths were positively correlated at this period, because high soil moisture in March and early April and after fall rains had begun was not accompanied by effective plant control. For the 13 increasingly effective spraying dates at the early part of the growth cycle (Table 5), total phloem thickness, rate of new xylem ring growth, upward rate of methylene blue dye movement, minimum and maximum leaf moisture stress, mean air temperature, and soil temperatures were directly correlated with plant control. Percent soil moisture at 61 cm was inversely correlated with control. Soil temperature at a depth of 91 cm was most correlated with control, and this finding supports the work of Dahl et al. (2) who found that the percentage of plants killed in June by 2,4,5-T was best correlated with increasing soil temperature at depths of 46 and 91 cm in northwest Texas. Apparently after the soil temperature reaches a threshold of 18° to 20° C, honey mesquite actively initiates and undertakes stem elongation growth, foliage production, and some new phloem and xylem radial growth. Both minimum and maximum leaf moisture stress were.- increasing, and soil moisture at 61 cm was decreas- ing at this period but did not limit plant control. For the 22 decreasingly effective dates from May 21 through October 9 (Table 5), plant control was directly correlated with rate of new xylem ring growth, rainfall 1 month before spraying, and per- cent soil moisture at 91 cm. Minimum leaf moisture stress and total new xylem ring thickness were in- versely correlated with plant control. Rate of new xylem radial growth was decreasing, while the total new xylem ring thickness was increasing, especially during the period from late May through lune. Minimum leaf moisture stress was increasing rapidly as rainfall the month before spraying and soil mois- ture decreased. Upward methylene blue dye move- ment was not significantly correlated because the rate had fallen off faster than plant control. Of the plant variables, over all 36 dates, total and translocating phloem essentially were only cor- related with each other (0.73). Total new xylem ring 1 thickness increased to a maximum in late June and was closely correlated with increasing leaf moisture stress, rising air and soil temperatures, and decreas- ing soil moisture (Table 6). The rate of new xylem ring growth was positively correlated with increasing phloem thickness. As expected, increasing leaf moisture stess was accompanied by higher air and soil temperatures and lower soil moisture. Total available carbohydrates decreased to a minimum aboutApril and then increased again along with total new xylem ring thickness, maximum leaf moisture stress, mean air temperature, and soil temperature at a depth of 91 cm. For simple correlations of environmental var- iables at all 36 dates, mean air temperature the day 9 Table 5. Simple correlation coefficients between plant and environmental variables and percent canopy reduction (C.R.) by both herbi percent dead honey mesquite plants (D.P.) by picloram + 2,4,5-T 1 Time period2 13 Increasingly 22 Decreasingly ,_ 33 Dates3 effective dates4 effective dates5 2,4,5-T Picloram + 2,4,5-T Picloram + 2,4,5-T Piclora 2,4,5-T 2,4,5-T _ 2,4,5- y C.R. C.R. D.P. C.R. C.R. D.P. 0.28. C.R. ~ Variable (%) (%) (%) (%) (%) (%) (%) (%) Phloem thickness, mm Total .75 .79 .76 Translocating New xylem ring thickness, mm Total —.31 —.42 —.45 .61 —.49 —.45 Rate of radial growth .54 .61 .55 .67 .74 .59 .66 .71 Leaf moisture stress, bars Minimum —.59 —.63 —.44 .60 .66 .77 —.71 —.78 Maximum .73 .77 Total available carbohydrates, % Upward methylene blue Dye movement, cm/hr .49 .49 .58 .70 .71 .75 Mean air temperature, 0C Day of spraying -—.43 —.36 —.44 .74 .78 1 wk before spraying —.37 —.36 —.33 .80 .68 Soil temperature, 0C 30 cm deep -.38 —.43 —.47 .81 .67 91 cm deep —.43 —.36 .82 .85 .87 Rainfall, cm 1 wk before spraying 1 month before spraying .41 .42 .33 .42 .45 Soil moisture, % 30 cm deep .37 .40 .34 61 cm deep .60 .58 .40 —.60 —.72 —.75 .46 91 cm deep .66 .62 .43 —.68 .43 .50 , 1C.Fl. = Canopy reduction; D.P. = Dead plants. 233 Dates = April 29-August 31; 13 increasingly effective dates = March 24-May 29; 22 decreasingly effective dates = May 21-October 3Correlation coefficients of i033 or i044 are significant at the 5% and 1% level, respectively. "- 4Correlation coefficients of i054 or 10.68 are significant at the 5% and 1% level, respectively. l sCorrelation coefficients of £0.42 or i053 are significant at the 5% and 1% level, respectively. of spraying was directly correlated with mean air temperature 1 week before spraying (0.87, signifi- cance at 5%/1% levels = i0.32/i0.42, respectively). Soil temperatures at both depths were directly and highly correlated with each other and mean air tem- perature (0.79 to 0.87). Rainfall 1 week and 1 month before spraying were directly but weakly correlated with each other (0.46) and inversely correlated with mean air temperature (—0.34 to —0.44) and soil tem- perature (—0.38 to —0.51). Soil moisture levels at all three depths were directly and highly correlated with each other (0.79 to 0.94) and generally with rain- fall (0.40 to 0.64) and were inversely correlated with mean air temperature (—0.62 to —0.72) and soil tem- perature (—0.72 to ,—0.78). This again reflects the warming and drying trend throughout the growing season. Regression equations were developed from the data of all 4 years to compute the best estimate (Y) for the observed percent canopy reduction and per- 10 cent dead honey mesquite plants resulti ‘ herbicide applications at any time duringt j ing season. Hopefully, these equations wi fit data from similar treatments showing Q honey mesquite responses to 2,4,5-T and pi 2,4,5-T. lf reliable equations can be develop can be used commercially to predict the a l honey mesquite control that can be expeci the treatment under any given seasonal growth conditions. ,7 Simple regression equations were calc I predicting (Y) percent canopy reduction f, and picloram + 2,4,5-T and percent plants . picloram + 2,4,5-T at four intervals duringt j study period (Table 7). An insufficient n plants was killed by 2,4,5-T to calculate a si ,1 equation for percent dead plants. At all from March 24 to October 9, rate of new xyl growth gave the best equation for all t sponses to herbicides. Apparently the i; " - 6. Simple correlations of selected honey mesquite plant variables with other plant and environmental variables at 36 dates between March F d October 9 over a 4-year period at Bryan and Millican, Texas New xylem Leaf moisture rmg thwkness stress Methylene blue Total available Total Rate Minimum Maximum dye movement carbohydrates gl ble (mm) (mm) (bars) (bars) (cm/hr) (%) m thickness, mm tal .56 .44 ‘_ nslocating .43 .34 ‘ xylem ring thickness, mm tal .63 .71 —.44 .51 te of radial growth —.44 .45 moisture stress, bars 'nimum .60 —.37 ximum .39 g available carbohydrates, % —.42 ‘ air temperature, C of spraying .82 .48 .77 .41 = k before spraying .79 .54 .70 —.38 .41 temperature, C I cm deep .78 .67 .81 —.42 _ cm deep .75 .62 .71 .34 A fall, cm wk before spraying -—.35 —.44 month before spraying —.41 —.59 —.60 ‘ oisture, % -= cm deep —-.41 .32 —.59 -—.58 i. cm deep —.59 —.71 —.63 cm deep —.64 —.75 —.56 .41 relation of 10.32 or i0.42 are significant at the 5% and 1% level, respectively. i 7. Simple regression equations for predicting response of honey mesquite near Bryan and Millican, Texas, to herbicides with plant and nmental variables during four seasonal intervals over a 4-year period lde Type control1 Equation R2 * All 36 dates between March 24 and October 9 I T % C.Fi. i= 28.6 + 123.8(rate of new xylem growth) 0.45 am + 2,4,5-TL % C.R. /Y\ = 36.2 + 188.0(rate of new xylem growth) .52 % D.P. Y = 3.18 +164.9(rate of new xylem growth) .39 33 dates between April 29 and August 31 % on. jf= 5.54 + 3.08m soil moisture, 91 cm) .44 m + 2,4,5-T % C.R. X = 137.9 — 6.36(minimum leaf moisture stress) .40 % D.P. Y = 13.7 + 0.20(upward methylene blue dye movement) .33 . A 13 increasingly effective dates between March 24 and May 29 T % C.R. /Y\= —25.7 + 4.09(soil temperature, 91 cm) .67 am + 2,4,5-T % CR. Y = 48.3 + 6.14(soil temperature, 91 cm) .73 % D.P. Y = —83.1 + 63.5(soil temperature, 91 cm) .75 /2\ decreasingly effective dates between May 21 and October 9 T % C.Fl. ix= 108.2 — 5.64(minimum leaf moisture stress) .50 m + 2,4,5-T % C.R. X = 162.2 — 8.85(minimum leaf moisture stress) .60 % D.P. Y = —0.55 + 165.6lrate of new xylem growth) .46 .Fl. = % Canopy reduction; % D.P. = % Dead plants. quations are significant at the 1% level. g carbohydrates stored in the stem and roots for For 33 dates between April 29 and August 31, al growth. Therefore, applied herbicides, which when the plants were generally fully foliated, the w carbohydrate usage patterns, would be ex- best equations pertained to moisture. Canopy re- ed to be translocated rapidly to the stern and duction by 2,4,5-T was best predicted by percent soil s where they could kill the plants. moisture at a depth of 91 cm. For picloram + 2,4,5-T, ‘l1 canopy reduction and dead plants were best pre- dicted by minimum leaf moisture stress and rate of upward methylene blue dye movement, respective- ly. The early spring treatment dates were highly correlated with soil temperature at the depth of 91 cm. This occurred because of the steady increase in temperature along with plant control. However, in late summer, air and soil temperatures remained high, but plant control decreased markedly. Appar- ently a minimum temperature of about 20° C is re- quired for honey mesquite to grow. Above this threshold temperature, growth processes increased up to a point. Later in the summer, as shown by the decreasing 22 dates, moisture stress appeared to be the major limiting factor for controlling honey mes- quite. Even as late as October, air and soil tempera- tures were warm. Apparently at this time, rainfall and soil moisture usually decrease to a point where the plants enter a high period of stress. In 1969, however, about 25 cm of rain fell during the month before the October 9 spraying. This rain increased the soil moisture at a depth of at least 91 cm, but failed to activate the plant growth even though temperatures were favorable. Apparently these honey mesquite plants were deep-rooted and extracted a large percentage of their water from deep in the soil. Table 8. Three-factor multiple regression equations for predicting response of honey mesquite near Bryan and Millican, Texas, to ’ with plant and environmental variables during four seasonal intervals over a 4-year period Multiple regression equations with thr iables were calculated to predict percent can duction by 2,4,5-T and picloram + 2,4,5-T an cent plants killed by picloram + 2,4,5-T for" dates, 33 dates between April 29 and August, increasingly effective early dates between and May 29, and 22 decreasingly effective da ' tween May 21 and October 9 (Table 8). 1 For all 36 dates, rate of newxylem radial g‘ percent total available carbohydrates, and l‘ soil moisture at the depth of 30 cm were th important plant and environmental factors prediction of canopy reduction by both her However, the T-T’ factors, to account for a pa 1 curve, coupled with rate of upward move a, methylene blue dye was a better predictive nation for picloram + 2,4,5-T. For predicti, percentage of plants killed by picloram + l rate of new xylem radial growth, rate of movement of methylene blue dye, and transl. phloem thickness were most important. 1 For 33 dates between April 29 and Aug,‘ (Table 8), rate of new xylem radial growth and. locating phloem thickness were important if, dicting canopy reduction for both herbicid. T-T’ factors gave a better equation when c. with translocating phloem thickness than th plant variables for predicting percent canopy Herbicide Type controll Equation A|lA36 dates between March 24 and October 9 2,4,5-T % C.R. Y = 90.8 + 161 .5(X1l —— 2;33(X2l—1.00lX3l ePicloram + 2,4,5-T % CR. A % C.R. x % D.P. Y = 33 2,4,5-T % C.R. ¢= Picloram + 2,4,5-T % C.R. >t= % C.R. :/\= % D.P. Y = 13 increasingly 2,4,5-T % 0R. i = Picloram + 2,4,5-T % C.R. x % D.P. Y = 2,4,5-T % C.Fi. A Picloram + 2,4,5-T % C.Fi. /Y\ = % D.P. Y % D.P. '\?= 9 = 93.3 + 22s.4(x1) —1.88lX2l—1.29lX3l = -1e4.0 + 2.80m - 0.00331?) + 0.1 161x41 6.45 +140.9lX1) + 0.16lX4) —— 91 .9(X5l dates between April 29 and August 31 10.6 + 59.2(X1l — 64.4(X5l‘+ 2.30(X6) 60.4 + 122.1 (X1) + O.10(X4l — 97.7(X5l -111.4 + 2.79m - 0.0091 (T2) - 4e.5lxs) -51.5 + 3.55(X2l + 0.32(X4) —147.3lX5l effective early dates between March 24 and May 29 —160.6 —1.40(X2l+ 5.43(X6l + 5.98(X7l = —112.6 +1.94(X3l+ 5.66(X7l+1.77(X8l —51.3 +1.14(X3)+10.2(X7l — 5.84(X9l 22 decreasingly effective dates between May 21 and October 9 Y = —34.5 + 77.1 (X1) + 4.31 (X9) —— 5.25(X10l —14.6 +103.8(X1l+ 5.15lX9l — 8.00(X10) = 53.3 +188.9(X1l —122.6(X5)— 25.6(X11l 543.1 - 7.05m + 0.015202) + 9181x121 lType control are % C.R. = % Canopy reduction; % D.P. = Dead plants. *Variable abbreviations are the following: T = time in days from January 1 to spraying date; T2 = time period (Tl squared; X1 = _ xylem ring radial growth; X2 = percent total available carbohydrates; X3 = percent soil moisture at the depth of 0 to 30.5 cm; X upward movement of methylene blue dye in the xylem; X5 = translocating phloem thickness; X6 = percent soil moisture at the de _ to 91.4 cm; X7 = soil temperature at a depth of 91.4 cm; X8 = maximum daily leaf moisture stress; X9 = soil temperature at a =- cm; X10 = minimum daily leaf moisture stress; X11 = total new xylem ring radial growth; X12 = mean air temperature 1 week befor 1 The T, T2 factors are presented only where they both occurred in the equation. All equations are significant at the 1% level. 12 q by picloram + 2,4,5-T. Percent total available bohydrates, rate of upward movement of thylene blue dye, and translocating phloem l For the 13 increasingly effective early dates be- en March 24 and May 29 (Table 8), soil tempera- at a depth of 91 cm occurred in all equations, as j asa soil moisture factor. The R’ values at this A- period were the highest of the four periods ied. I For the 22 decreasingly effective dates between y, 21 and October 9 (Table 8), rate of new xylem ~ radial growth occurred in three equations. Soil Vperature at a depth of 30 cm and minimum leaf isture stress occurred in canopy reduction Vations for both 2,4,5-T and picloram + 2,4,5-T. T + T2 factors, indicating a hyperbolic curvilinear tionship, accompanied by mean air temperature eek before spraying gave the best predictive equa- 1 for the percentage of plants killed by picloram + ,5-T. This equation was better than the one with three plant variables including rate of new xylem ‘ radial growth, thickness of translocating em, and thickness of the new xylem ring. In summary, the most effective period for spray- honey mesquite with these two herbicides was from late April through June which confirms other research and applied knowledge. The mixture of pic- loram + 2,4,5-T was more effective than 2,4,5-T alone. The plant and environmental variables most corre- lated with control varied depending on the seasonal period considered. For the period between April 29 and August 31, when plants were almost fully foliated, control was most correlated with soil moisture level at a depth of 91 cm, minimum leaf moisture stress, or rate of upward methylene blue dye movement. During the early spring growth period, control was most correlated (positively) with soil temperature at a depth of 91 cm when generally adequate moisture was present. Soil temperature is one of the easiest variables to measure and presum- ably could be used on a practical basis. Later in the season when temperature was adequate, moisture, as measured by daily minimum leaf moisture stress, became the most limiting factor (negatively) affecting at least percent canopy reduction. Therefore, a com- bination of warm temperature and adequate plant moisture content is necessary for effective control with the herbicides used. These results are applicable to clay soil sites at least in East Texas. However, the relative importance of these factors may change for plants growing on other soil types or in different climatic conditions. 13 LITERATURE CITED . Brady, H. A. 1971. Spray date effects on behavior of herbicides on brush. Weed Sci. 19:200-202. . Dahl, B. E., R. B.Wadley, M. R. George,andJ. L.Talbot.1971, lnfluence of site on mesquite mortality from 2,4,5-T. J. Range Manage. 24:210-215. . Davis, F. S., R. E. Meyer, J. R. Baur, and R. W. Bovey. 1972. Herbicide concentrations in honey mesquite phloem. Weed Sci. 20:264-267. . Fisher, C. E., J. L. Fults, and H. Hopp. 1946. Factors affecting action of oils and water-soluble chemicals in mesquite eradica- tion. Ecol. Monogr. 16:100-126. . Fisher, C. E., C. H. Meadors, R. Behrens, E. D. Robison, P. T. Marion, and H. L. Morton. 1959. Control of mesquite on graz- ing lands. Tex. Agr. Exp. Sta. Bul. 935. 24 pp. . Fisher, C. E., E. D. Robison, G. O. Hoffman, C. H. Meadors, and B. T. Cross. 1970. Aerial application of chemicals for con- trol ofbrush on rangelands. Tex. Agr. Exp. Sta. Prog. Rep. 2801. ' pp 5-11. 14 . Flynt, T. 0., T. E. Riley, R. W. Bovey, and R. E. Meyer. 1971. Auger soil sampler for herbicide residues. Weed Sci. 19:583- 584. . Haas, R. H., and J. D. Dodd. 1972. Water-stress patterns in honey mesquite. Ecol. 53:674-680. APPENDIX English Units Bar Centigrade (C) Meter (rn) Millimeter (mm) Centimeter (cm) Kilometer (km) Hectare (ha) Gram (g) Kilogram (kg) Kilogram per hectare (kg/ha) Liter (L) This publication reports research involving pesticides. It does not contain recommendations for their use, nor does is imply that the uses discussed here have been registered. All uses of pesticides must be registered by appropriate State and/or Federal agencies before they can be recommended. 9. 10. 11. 12. 13. 14. 15. McMillan, Calvin, and J. T. Peacock. 1964. Bud-bu diverse populations of mesquite (Prosopis: Legu y under unform conditions. Southwest Nat. 9:181-188. Meyer, R. E., R. W, Bovey, W. T. McKelvy, and T. ’ 1972. Influence of plant growth stage and environm_ tors on the response of honey mesquite to herbiciy Agr. Exp. Sta. Bul. 1127. 19 pp. f. j Meyer, R. E., R. w. Bovey, T. E. Riley,“ and w. T. ‘ 1972. Leaf removal interval effect after sprays to Weed Sci. 20:498-501. y‘ Meyer, R. E., R. H. Haas, and C. W. Wendt. 1972. lnt of environmental variables and growth and develo’, honey mesquite. Bot. Caz. 134(3)’173-178. Meyer, R. E., H. L. Morton, R. H. Haas, E. D. Rob’ T. E. Riley. 1971. Morphology and anatomy of ho ' quite. USDA Tech. Bul. 1423. 186 pp. F Robison, E. D., R. E. Meyer, B. T. Cross, and H. L.‘ 1970. Influence of preconditioning defoliations _ mesquite control. Tex. Agr. Exp. Sta. Prog. Rep. 2808. _ Wilson, R. T., B. E. Dahl, and D. R. Krieg. 1975. Carl concentrations in honey mesquite roots in rel phenological development and reproductive con Range Manage. 28:286-289. ‘ Metric Units 0.99 Atmosphere 5/9 (Fahrenheit — 32) 3.28 Feet 0.039 Inch 0.39 Inch 0.62 Mile 2.47 Acres 0.0022 Pound 2.2 Pounds 0.89 Pound per acre 1.06 Quarts TEXAS AGRICULTURAL EXPERIMENT STATION MAIN STATION DEPARTMENTS COLLEGE OF AGRICULTURE COLLEGE OF VETERINARY MEDICINE Agricultural Analytical Services Veterinary Research-General Agricultural Communications Veterinary Medicine and Surgery Agricultural Economics Veterinary Microbiology Agricultural Engineering Veterinary Parasitology Animal Science Veterinary Pathology Biochemistry and Biophysics Veterinary Physiology and Pharmacology Consumer Research Center Veterinary Public Health Entomology . Institute of Tropical Veterinary Medicine Feed and Fertilizer Control Service Forest Science Horticultural Sciences Plant Sciences Poultry Science Range Science Recreation and Parks Rural Sociology Soil and Crop Sciences Wildlife and Fisheries Sciences AGRICULTURAL RESEARCH UNITS Texas A&M University Agricultural Research and Extension Center at AMARILLO (Bushland) Texas A&M University Agricultural Research and Extension Center at BEAUMONT Western Division at EAGLE LAKE Texas A&M University Agricultural Research Station at ANGLETON Texas A&M University Agricultural Research and Extension Center at CHILLICOTHE-VERNON Texas A&M University Agricultural Research Station at CHILLICOTHE Texas A&M University Research Station at IOWA PARK Texas A&M University Vegetable Station at MUNDAY Texas A&M University Agricultural Research Station at SPUR Texas Experimental Ranch, THROCKMORTON Texas A&M University Agricultural Research and Extension Center at CORPUS CHRISTI Texas A&M University Agricultural Research Station at BEEVILLE Texas A&M University Research and Extension Center at DALLAS Texas A&M University Research Center at EL PASO Texas A&M University Agricultural Research Station at PECOS Texas A&M University Agricultural Research and Extension Center at LUBBOCK High Plains Research Foundation (Halfway) Texas A&M University-Texas Tech University Cooperative Research Unit at LUBBOCK Texas A&M University Agricultural Research Center at McGREGOR Texas A&M University Agricultural Research and Extension Center at OVERTON Prairie View-Texas A&M University Research Center at PRAIRIE VIEW Texas A&M University Agricultural Research and Extension Center at SAN ANGELO Texas A&M University Agricultural Research Station at SONORA Texas Range Station, BARNHART Texas A&M University Agricultural Research and Extension Center at STEPHENVILLE Texas A&M University Fruit Research — Demonstration Station at MONTAGUE Blackland Research Center at TEMPLE Texas A&M University Agricultural Research and Extension Center at UVALDE Texas A&M University Agricultural Research and Extension Center at WESLACO Texas A&M University-Texas A&l University Cooperative Research Unit at WESLACO Texas A&M University Plant Disease Research Station at YOAKUM TEXAS WATER RESOURCES INSTITUTE The Texas Agricultural Experiment Station J. E. Miller, Director, College Station, Texas 2.5M—1-77 a All programs and information of The Texas Agricultural Experiment Station are available to everyone without regard to race, color, re- ligion, sex, age, or national origin.